- Email: [email protected]
An experimental investigation of the impact of the Lidcombe Program on early stuttering
Journal of Fluency Disorders 27 (2002) 203–214
An experimental investigation of the impact of the Lidcombe Program on early stuttering Vanessa Harris a , Mark Onslow b,∗ , Ann Packman b , Elisabeth Harrison a , Ross Menzies c a
Stuttering Unit, Bankstown Health Service, Sydney, Australia Australian Stuttering Research Center, The University of Sydney, P.O. Box 1825, NSW 2141 Lidcombe, Australia School of Behavioral and Community Health Sciences, The University of Sydney, Sydney, Australia b
c
Received 25 September 2001; received in revised form 12 March 2002; accepted 14 March 2002
Abstract Preliminary Phase I and II trials for the Lidcombe Program of early stuttering intervention have found favorable outcomes and that the treatment is safe. Although speech–language pathologists (SLPs) often need to intervene with pre-schoolers’ early stuttering, many of these children will recover at some time in the future without such intervention. Consequently, they need to know whether the Lidcombe Program’s effect on stuttering is greater than that of natural recovery. Participants were 23 pre-school children who were randomly assigned to either a control group or a treatment group that received the Lidcombe Program for 12 weeks. A repeated measures ANOVA showed no main effect on stuttering for the group (control/treatment), a significant main effect for the measurement occasion (at the start and at the end of the treatment period), and a significant interaction between group and measurement occasion. Stuttering in the treatment group reduced twice as much as in the control group. These results are interpreted to mean that the introduction of the Lidcombe Program has a positive impact on stuttering rate, which exceeds that attributable to natural recovery. Educational objectives: Readers will learn about and be able to describe: (1) how natural recovery can affect assessments of the effectiveness of treatments for early stuttering; (2) the relative effects of the Lidcombe Program and natural recovery on stuttering; and
∗
Corresponding author. Tel.: +61-2-9351-9061; fax: +61-2-9351-9392. E-mail address: [email protected] (M. Onslow).
0094-730X/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 9 4 - 7 3 0 X ( 0 2 ) 0 0 1 2 7 - 4
204
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
(3) the difference between the results of this study and those of uncontrolled clinical trials. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Childhood stuttering; Lidcombe Program; Short-term effects
1. Introduction The Lidcombe Program is a behavioral treatment for stuttering in pre-school-age children. The treatment is conducted by parents in the child’s everyday environment and parents learn how to do the treatment during the weekly visits with the child to the speech–language pathologist (SLP). A comprehensive description of the Lidcombe Program can be found in Onslow, Packman, & Harrison (in press). In the Lidcombe Program, the parent gives verbal contingencies during conversational exchanges with the child. These verbal contingencies are directed at: (1) stutter-free speech; (2) unambiguous stuttering; (3) correct self-evaluation of stutter-free speech; and (4) spontaneous self-correction of stuttering. These verbal contingencies consist of: (1) acknowledgment and/or praise for periods of stutter-free speech; (2) acknowledgment of stuttering and/or a request that the child corrects stuttering; (3) praise for correct self-evaluation of stutter-free speech; and (4) praise for spontaneous self-correction of stuttering. The SLP ensures that these parental verbal contingencies are not constant, intensive or invasive, and that parents are at all times positive and supportive of the child receiving the treatment. The treatment is individualized for each family and, as with any treatment for a childhood speech or language disorder, it is essential that the child enjoys the treatment and finds it to be a positive experience. Stuttering measures are an essential component of the Lidcombe Program. The parent makes daily measures of the severity of the child’s stuttering on a 10-point scale, where 1 = no stuttering; 2 = very mild stuttering; and 10 = extremely severe stuttering. The SLP makes weekly measures of stuttering rate (percent syllables stuttered, %SS). Together, these two measures are used to: (1) guide implementation of the program from week to week; (2) identify when the child has met criterion speech performance; and (3) check that the child’s speech continues to meet criterion speech performance in the long-term. The stuttering measures also enable the SLP and the parent to communicate effectively about the severity of the child’s stuttering throughout the treatment process. The Lidcombe Program is conducted in two stages. Stage 1 is complete when the child’s stuttering is below 1.0%SS and each of the daily severity ratings for the corresponding week are either 1 or 2, with the majority being 1. During Stage 2, the parent gradually withdraws the verbal contingencies and gradually assumes complete responsibility for the treatment as visits to the clinic decrease in frequency. Any departure from the criterion speech performance, as specified with the stuttering measures at the end of Stage 1, results in more frequent clinic visits and possibly an increase in parental contingencies.
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
205
A file audit of 250 children treated with the Lidcombe Program (Jones, Onslow, Harrison, & Packman, 2000) showed that the mean number of weekly clinic visits taken to complete Stage 1 was 12.5, and a recovery plot for these children showed that 90% of them had completed Stage 1 after 22 clinic visits. Jones et al. found that pre-treatment stuttering rate was the only predictor of time required to complete Stage 1. Other case variables — most notably time since reported onset — did not affect the time taken to complete Stage 1. The safety and clinical promise of the Lidcombe Program have been established in Phase I and II (Pocock, 1996) trials. Non-controlled trial data (Onslow, Andrews, & Lincoln, 1994; Onslow, Costa, & Rue, 1990) and file audit data (Jones, Onslow, Harrison, & Packman, 2000) have shown outcomes of zero or near zero rates of stuttering in children in the medium-term after treatment with the Lidcombe Program. Long-term data on 42 children (Lincoln & Onslow, 1997) show outcome to be durable. The social validity of treatment outcomes has been established (Lincoln, Onslow, & Reed, 1997). Given that the Lidcombe Program is a direct treatment, it is perhaps not surprising that concerns have been expressed about its safety (Cook, 1996; Cook & Rustin, 1997). Consequently, Woods, Shearsby, Onslow, and Burnham (2002) demonstrated that the treatment was not associated with psychological harm to children. To date, however, the efficacy of the Lidcombe Program has not been established. In other words, there is no unequivocal evidence that the reductions in stuttering that occur with the Lidcombe Program are in fact due to the treatment. The most important issue to consider in determining the effectiveness of treatments for early stuttering is, of course, natural recovery. Here, there are two questions (Packman & Onslow, 1998). The first is whether recovery from stuttering with the Lidcombe Program is greater than natural recovery in the long-term. In other words, after a clinically significant post-treatment period, are the improvements in stuttering that occur after treatment with the Lidcombe Program greater than those of natural recovery? The definitive answer to this question would involve comparing a group of children who receive no treatment with a group of children who receive treatment with the Lidcombe Program. In light of the fact that natural recovery may take 2 years or longer (e.g., Ingham & Riley, 1998; Yairi & Ambrose, 1999), this would involve withholding treatment for a control group of children for a number of years. This cannot be justified on ethical grounds, so the next best thing is to compare the Lidcombe Program with another treatment or treatments. Consequently, a Phase III (Pocock, 1996) randomized controlled trial of the program is in progress in New Zealand (Jones, Gebski, Onslow, & Packman, 2001). A second, and equally important question about effectiveness concerns the immediate effects of this treatment: When the Lidcombe Program is introduced, are its effects on the course of the disorder greater than the effects of natural recovery? Natural recovery occurs in many cases of early stuttering, with two recent estimates lying in the mid-70% range (Mansson, 2000; Yairi & Ambrose, 1999). At present, however, there is some doubt that recovery rates for children who present to a clinic would be as high as those population estimates. In any
206
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
event, in many instances SLPs will want to implement treatment sometime during the pre-school years, despite not knowing whether the child is destined to recover naturally. There are many reasons for clinicians to treat a pre-school child who stutters, even though there is a chance that natural recovery may occur at some future time (Packman & Onslow, 1999; Packman, Onslow, & Attanasio, in press). Consequently, there is need for evidence that the immediate reductions in stuttering observed with the Lidcombe Program can be attributed primarily to the treatment rather than to natural recovery. An experimental design is required to establish this evidence. Such a design involves withholding treatment for a control group, but only for a short period, and would determine whether the Lidcombe Program provides short-term benefits beyond what might be expected from natural recovery. The present paper reports such an experimental investigation. In this study, a group of children who receive the Lidcombe Program over 12 clinic visits is compared with a control group of children who receive no treatment for 12 weeks. Twelve clinic visits was chosen as this is close to the mean number of clinic visits (12.5) taken by pre-school children to complete Stage 1 of the Lidcombe Program (see earlier). Of course, not all children receiving the Lidcombe Program in the study would be expected to complete Stage 1 within 12 clinic visits, but this was considered a sufficient period to demonstrate the impact of the treatment on the natural course of stuttering.
2. Method 2.1. Participants The participants were 23 children assessed at the Stuttering Unit, Bankstown Health Service, and Campbelltown Hospital, Macarthur Health Service, in Sydney. The children lived in a range of suburbs in metropolitan Sydney of varying socio-economic levels. All children came to the clinic originally for stuttering, except for one child in the experimental group who presented with mild speech and language problems and who was found at assessment to be stuttering. At the conclusion of the experiment, one control child received additional treatment for language and voice problems, and one experimental child received additional treatment for articulation and language. The post-experiment treatment history of one experimental child and one control child was not available. Potential subjects were screened against selection criteria, which were: (1) age between 2 years 0 months and 4 years 11 months; (2) no previous treatment with the Lidcombe Program or professional consultation involving the Lidcombe Program; (3) 6 months or longer since reported onset of stuttering; (4) English the first language for both parents and child; and (5) stuttering at a rate of 3.0%SS or greater within the clinic during a conversation with the first investigator or a parent. For criterion (5), stuttering rate in %SS was obtained from an audio recording of a within-clinic speech sample recorded while the child played in the clinic at the
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
207
assessment interview. Potential subjects at the Stuttering Unit were recruited in the order in which parents telephoned to request treatment for their child. Potential subjects from Campbelltown Hospital were recruited in order from a list of children waiting for assessment. Potential subjects were assessed during an interview with the first investigator. 2.2. Procedure Parents of children who met selection criteria at the assessment interview were informed about the study and asked if they wished to participate. Parents were told that: (1) they would be randomly allocated to either an experimental (treatment) group or a control (non-treatment) group; (2) participation in the study would last approximately 12 weeks; and (3) at the completion of the study, children allocated to the treatment group would continue on the program and children in the control group would receive treatment when their name came to the top of the clinic waiting list. Parents who agreed to participate were supplied with a small audiotape recorder and instructions to make 10 min beyond-clinic recordings in three situations during the following week. In these situations the child was to speak: (1) with a parent; (2) with a familiar person other than a parent, away from home; and (3) at any location with a familiar person without the child’s knowledge of the recording. At the assessment interview, information about the nature of stuttering was provided to parents but no detailed information was provided about either the Lidcombe Program or any other stuttering treatment. Parents were advised only that the Lidcombe Program is presented by parents and involves weekly visits to the speech clinic. An appointment was made for the following week and parents were advised that their child would be randomly allocated to either the treatment or the control group when they returned at that time with all three tape recordings completed. No advice was given to any parent about how to react to or otherwise deal with the child’s stuttering. Children were randomly allocated to either the treatment group or the control group at the beginning of the first clinic visit after the interview assessment visit, provided parents had made and produced the three beyond-clinic recordings. Randomization was performed independently at the Clinical Trials Center of the National Health and Medical Research Council of Australia. The first investigator telephoned the Clinical Trials Center, where the child was randomized using a blocked randomization design. As this was an experiment and not a clinical trial, an a priori decision was made that randomization would continue until there were at least 10 subjects in each group, even if that resulted in uneven group numbers. Subject numbers were not based on power calculations, because, as is often the case with a low incidence disorder such as stuttering (Jones, Gebski, Onslow, & Packman, 2002), requisite numbers of clinical subjects for adequate power were prohibitive in terms of feasibility for an initial study. In the present case, greater than 60 subjects per group would be required to detect a medium effect at power
208
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
0.8 and α 0.05. Instead, subject numbers were chosen so that the present study could serve to determine the viability of a larger scale project in the event that data were non-significant in the presence of apparent trends (Jones, Gebski, Onslow, & Packman, 2002). Children allocated to the treatment group began their 12 clinic visits immediately. For these children, one within-clinic recording and three beyond-clinic recordings were again made, in the same speaking situations, after the requisite 12 clinic visits. The parents of the control group children were scheduled to return to the clinic 12 weeks post-randomization. During this visit, a within-clinic recording was made, and the parents presented the required set of three beyond-clinic recordings made during the previous week. 2.3. Dependent measure The dependent measure was %SS. In contrast to definition-based measures of stuttering, the measure of %SS records the number of speech events thought by an experienced clinician to be unambiguous stuttering (Jones, Gebski, Onslow, & Packman, 2001). The %SS does not include counts of normal disfluencies. Measures of %SS were made by an SLP experienced in treating and measuring stuttering, who was independent of the study and had no knowledge of the participants. The clinician knew the topic of the research but not its details. Recordings were presented blind, in random order, to this clinician with pre-treatment and post-treatment status also randomized. The mean %SS scores for each child at 1 week pre-randomization (first measurement occasion) and at 12 weeks post-randomization (second measurement occasion) were calculated, based on the four tape recordings (three made beyond the clinic and one made within the clinic). Pre- and post-randomization within-clinic recordings for all children were selected for intrajudge reliability analysis. These 46 recordings, comprising around a quarter of the total recordings, were presented to the SLP for re-measurement, 1 month after the original measures were made. Interjudge reliability was established by a second SLP, who was also experienced in treating and measuring stuttering, and who was also independent of the study and had no knowledge of the participants. This clinician also knew the topic of the research but not its details. This SLP also made blind measurements from randomly arranged recordings. It was determined a priori that around one-third of the recordings would be included in the interjudge reliability analysis. Tapes were selected randomly from the pool of recordings and presented to the SLP for measurement.
3. Results Twenty-nine children were randomized to the experiment. Of these, six children did not complete the experimental procedures, for the following reasons: death in
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
209
Table 1 Number, gender, mean %SS and mean time since onset for the control and the treatment groups Number
Control Treatment
Boys
Girls
11 8
2 2
Mean %SS pre-randomization
Mean time since onset (months)
8.4 8.6
10 12
the family, failure to comply with experimental procedures, treatment sought closer to home and loss of contact with the family. This left 10 children in the treatment group and 13 in the control group. Information about the children is shown in Tables 1 and 2.
Table 2 The columns display, for individual children, gender, reason for individual referral, treatment subsequent to the experiment for communication disorders other than stuttering, months since reported onset, and reported history of stuttering in first- or second-degree relatives Subject
Group
Gender
Original referral
Subsequent treatment
Months since reported onset
Reported family history
1 2
Control Control
M M
Stuttering Stuttering
9 12
Unknown Yes
3 4 5 6 7 8 9 10 11
Control Control Control Control Control Control Control Control Control
M M M F M M M M M
Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering
10 6 12 12 12 12 6 12 12
Yes Unknown Unknown No Yes Yes Yes Yes Yes
12 13 14 15
Control Control Experimental Experimental
F M M M
Stuttering Stuttering Stuttering Stuttering
6 9 12 12
Yes Yes Yes Yes
16
Experimental
M
6
Yes
17 18 19 20 21 22 23
Experimental Experimental Experimental Experimental Experimental Experimental Experimental
F M M M M M F
Mild articulation/ language Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering
No Language/ voice No No No No No No No No History unavailable No No No History unavailable Language
16 11 12 8 12 13 18
Yes Yes Yes Yes No Yes Yes
No No No No No No No
210
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
Of the 184 recordings scheduled in the experiment (23 children×4 recordings× 2 assessment occasions), 178 (96.7%) were obtained by the investigators and used for data collection. The mean duration of tape recordings of the children was 9.2 min. The mean number of syllables counted on the recordings during the collection of %SS scores was 448 (S.D. = 159; range, 93–959). In order to avoid multiple statistical comparisons within a small data set, and in order to minimize the effects of the variability of early stuttering on analysis procedures, the stuttering rates for each child pre- and post-randomization were based on means of four recordings. Therefore, mean %SS scores in the analysis were generally based on speech samples in excess of 1500 syllables. Of the 46 samples re-measured for intrajudge reliability, 37 (78.7%) differed by less than 1.0%SS, 46 (97.9%) differed by less than 2.0%SS, and 46 (100%) differed by less than 3.0%SS. The Pearson correlation between the original ratings and these ratings was 0.99. Of the 66 samples re-measured by the independent clinician for interjudge reliability analysis, 18 (27.3%) differed by less than 1.0%SS, 39 (59.1%) differed by less than 2.0%SS, and 55 (83.3%) differed by less than 3.0%SS. The Pearson correlation between the original ratings and these ratings was 0.83. The results of intrajudge and interjudge reliability analyses showed that overall measurement error for %SS was likely to be small in relation to the data trends detected in the study (see Fig. 1), and that data trends pre- to post-randomization would be detectable with the %SS measure. The mean %SS scores for the treatment and the control groups at the pre-randomization measurement occasion were 8.6 (S.D. = 5.2; range, 2.6–20.1) and 8.4 (S.D. = 2.5; range, 3.6–13.1) respectively, and at the post-randomization measurement occasions were 3.5 (S.D. = 2.8; range, 0.6–9.2) and 5.8 (S.D. = 3.6; range, 2.3–15.3), respectively. Nine of the 10 experimental children reduced their
Fig. 1. Mean %SS scores for the treatment and control group pre-randomization and postrandomization.
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
211
%SS score from pre- to post-randomization, with the mean change in those children being a 39% reduction of the pre-randomization %SS score. The remaining participant’s pre-randomization score increased by 16%. Nine of the 13 control children reduced their %SS score from pre- to post-randomization, with the mean change in those nine children being a 26% reduction of the pre-randomization %SS score. The other four control children increased their stuttering rates by 54, 20, 42, and 6% of the pre-randomization score. As expected, %SS data were skewed (Jones, Onslow, Harrison, & Packman, 2000). Normally distributed data are necessary for the calculation of variances in ANOVA. Hence, data were log transformed to produce a normal looking distribution and analyzed with a repeated measures ANOVA. There was no main effect for group (control/treatment; F = 3.13, P > 0.05). In other words, there was no difference between the groups when both pre- and post-randomization %SS scores were taken into account. There was a significant main effect for measurement occasion (pre- and post-randomization; F = 28.32, P < 0.001). This means that when %SS scores for both groups are combined, there was a significant decrease in stuttering from the first to the second measurement occasion. There was a significant interaction between group and measurement occasion (F = 5.02, P < 0.05). In other words, the treatment group improved significantly more than the control group from the first to the second measurement occasion. The significant interaction is displayed in Fig. 1.
4. Discussion The long-term effects of the Lidcombe Program are being studied in the Jones, Gebski, Onslow, & Packman (2001) clinical trial. The present study dealt with short-term effects, and showed that the improvement in stuttering in a group of pre-school-age children after a set number of clinic visits in the Lidcombe Program was greater than the effect of natural recovery. In other words, the results indicate that the treatment has an immediate impact on the natural course of the disorder. While some reduction in %SS scores occurred in the children who did not receive treatment, the decrease in stuttering in the treated group was around twice that amount. These results show that clinicians can be confident that implementing the Lidcombe Program will have a positive impact on stuttering rate in their pre-school clients. Stuttering frequency was the only dependent variable in the present study, so these results offer no explanation for why the treatment might be effective. No attempt was made to investigate reasons for the improvement in either group in this study, because measures of stuttering were made only after 12 weeks and before the completion of treatment, which is not a suitable design for such an investigation. Nonetheless, the reasons for recovery from stuttering during the Lidcombe Program is an interesting issue. At present, preliminary data suggest that neither extensive change in children and parents’ language habits (Bonelli, Dixon,
212
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
Bernstein Ratner, & Onslow, 2000), nor changes in speech acoustics (Onslow, Hewat, McLeod, & Packman, 2002), are associated with the treatment. The use of the present controlled experimental method would be an effective means for continuing this search for any changes in children or their parents that might explain the effects of the Lidcombe Program. The design of this experiment involved randomizing children, as they entered the study, either to the treatment group or the control group. This design, which is a true experimental design, was chosen in preference to a matched cohort design, which is a quasi-experimental design. The experimental design was chosen because it eliminates bias. It might be argued that children in a study such as the present one should be matched on variables such as stuttering rate, age, time since onset of stuttering, gender, and family history of recovery from stuttering. It could also be argued that they should be matched for cognitive, language and phonological development. Matching on so many variables would be unmanageable, and failure to match on any or all of them would leave the design open to the effects of bias. Even if matching on all such variables were to be achieved, there may be other variables, yet unknown to us, that might bias the results of such a study. The control children, as a group, showed some improvement in stuttering during the period of study, yet it is difficult to interpret this result considering that these are the first reported clinical measures of %SS over so short a period for a group of stuttering pre-school children who do not receive treatment. It may well be the case that some of the children in the control group who showed clinically significant stuttering reductions during the period of study may have continued that trend and recovered from the condition in a timely fashion. The present results raise the need to explore this issue in further clinical investigations. References Bonelli, P., Dixon, M., Bernstein Ratner, N., & Onslow, M. (2000). Child and parent speech and language and the Lidcombe Program of early stuttering intervention. Clinical Linguistics and Phonetics, 14, 427–446. Cook, F. (1996). The Lidcombe Programme: Is this the cure? Bulletin of the Royal College of Speech and Language Therapists, 528, 14. Cook, F., & Rustin, L. (1997). Commentary on the Lidcombe Program of early stuttering intervention. European Journal of Disorders of Communication, 32, 250–258. Ingham, J. C., & Riley, G. (1998). Guidelines for documentation of treatment efficacy for young children who stutter. Journal of Speech, Language, and Hearing Research, 41, 753–770. Jones, M., Gebski, V., Onslow, M., & Packman, A. (2001). Design of randomized controlled trials: Principles and methods applied to a treatment for early stuttering. Journal of Fluency Disorders, 26, 1–21. Jones, M., Gebski, V., Onslow, M., & Packman, A. (2002). Power in health research: A tutorial. Journal of Speech, Language, and Hearing Research, 45, 243–255. Jones, M., Onslow, M., Harrison, E., & Packman, A. (2000). Treating stuttering in children: Predicting treatment time in the Lidcombe Program. Journal of Speech, Language, and Hearing Research, 43, 1440–1450. Lincoln, M., & Onslow, M. (1997). Long-term outcome of an early intervention for stuttering. American Journal of Speech–Language Pathology, 6, 51–58.
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
213
Lincoln, M., Onslow, M., & Reed, V. (1997). Social validity of an early intervention for stuttering: The Lidcombe Program. American Journal of Speech–Language Pathology, 6, 77–84. Mansson, H. (2000). Childhood stuttering: Incidence and development. Journal of Fluency Disorders, 25, 47–57. Onslow, M., Andrews, C., & Lincoln, M. (1994). A control/experimental trial of an operant treatment for early stuttering. Journal of Speech and Hearing Research, 37, 1244–1259. Onslow, M., Costa, L., & Rue, S. (1990). Direct early intervention with stuttering: Some preliminary data. Journal of Speech and Hearing Disorders, 55, 405–416. Onslow, M., Hewat, S., McLeod, S., & Packman, A. (2002). Speech segment timing in children after the Lidcombe Program of early stuttering intervention. Clinical Linguistics and Phonetics, 16, 21–33. Onslow, M., Packman, A., & Harrison, E. (in press). The Lidcombe Program of early stuttering intervention: A clinician’s guide. Austin, TX: Pro-Ed. Packman, A., & Onslow, M. (1998). What is the take-home message from Curlee and Yairi? American Journal of Speech–Language Pathology, 7, 5–9. Packman, A., & Onslow, M. (1999). Issues in early intervention. In M. Onslow & A. Packman (Eds.), Handbook of early stuttering intervention. San Diego, CA: Singular Publishing Group. Packman, A., Onslow, M., & Attanasio, J. (in press). The timing of intervention with the Lidcombe Program. In M. Onslow, A. Packman, & E. Harrison (Eds.), The Lidcombe Program of early intervention: A clinician’s guide. Austin, TX: Pro-Ed. Pocock, S. J. (1996). Clinical trials: A practical approach. Chichester, England: Wiley. Woods, S., Shearsby, J., Onslow, M., & Burnham, D. (2002). The psychological impact of the Lidcombe Program of early stuttering intervention: Eight case studies. International Journal of Language and Communication Disorders, 37, 31–40. Yairi, E., & Ambrose, N. G. (1999). Early childhood stuttering I: Persistency and recovery rates. Journal of Speech, Language, and Hearing Research, 42, 1097–1112.
CONTINUING EDUCATION An experimental investigation of the impact of the Lidcombe Program on early stuttering QUESTIONS 1. This experiment was designed to determine: a. the long-term effects of the Lidcombe Program on stuttering b. the short-term effects of the Lidcombe Program on stuttering c. the long-term effects of natural recovery on stuttering d. the short-term effects of natural recovery on stuttering e. the relative effects of the Lidcombe Program and natural recovery on stuttering 2. The Lidcombe Program of early stuttering intervention is: a. a speech-pattern change procedure b. a procedure where demands on the child are reduced c. a behavioral treatment d. a cognitive-behavior treatment e. all of the above 3. In this experiment, the dependent variable was: a. the group to which the children were assigned b. percentage of syllables stuttered
214
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
c. stuttering severity measured on a 10-point scale d. syllables per minute e. “b” and “d” above 4. The design of this experiment was: a. a randomized controlled design b. a quasi-experimental design c. a crossover design d. a double-blind design e. none of the above 5. In this experiment, children in the treatment group reduced their stuttering: a. almost to the same extent as the control children b. twice as much as the control children c. half as much as the control children d. more than any other treatment group studied e. more than one of the above
An experimental investigation of the impact of the Lidcombe Program on early stuttering Vanessa Harris a , Mark Onslow b,∗ , Ann Packman b , Elisabeth Harrison a , Ross Menzies c a
Stuttering Unit, Bankstown Health Service, Sydney, Australia Australian Stuttering Research Center, The University of Sydney, P.O. Box 1825, NSW 2141 Lidcombe, Australia School of Behavioral and Community Health Sciences, The University of Sydney, Sydney, Australia b
c
Received 25 September 2001; received in revised form 12 March 2002; accepted 14 March 2002
Abstract Preliminary Phase I and II trials for the Lidcombe Program of early stuttering intervention have found favorable outcomes and that the treatment is safe. Although speech–language pathologists (SLPs) often need to intervene with pre-schoolers’ early stuttering, many of these children will recover at some time in the future without such intervention. Consequently, they need to know whether the Lidcombe Program’s effect on stuttering is greater than that of natural recovery. Participants were 23 pre-school children who were randomly assigned to either a control group or a treatment group that received the Lidcombe Program for 12 weeks. A repeated measures ANOVA showed no main effect on stuttering for the group (control/treatment), a significant main effect for the measurement occasion (at the start and at the end of the treatment period), and a significant interaction between group and measurement occasion. Stuttering in the treatment group reduced twice as much as in the control group. These results are interpreted to mean that the introduction of the Lidcombe Program has a positive impact on stuttering rate, which exceeds that attributable to natural recovery. Educational objectives: Readers will learn about and be able to describe: (1) how natural recovery can affect assessments of the effectiveness of treatments for early stuttering; (2) the relative effects of the Lidcombe Program and natural recovery on stuttering; and
∗
Corresponding author. Tel.: +61-2-9351-9061; fax: +61-2-9351-9392. E-mail address: [email protected] (M. Onslow).
0094-730X/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 0 9 4 - 7 3 0 X ( 0 2 ) 0 0 1 2 7 - 4
204
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
(3) the difference between the results of this study and those of uncontrolled clinical trials. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Childhood stuttering; Lidcombe Program; Short-term effects
1. Introduction The Lidcombe Program is a behavioral treatment for stuttering in pre-school-age children. The treatment is conducted by parents in the child’s everyday environment and parents learn how to do the treatment during the weekly visits with the child to the speech–language pathologist (SLP). A comprehensive description of the Lidcombe Program can be found in Onslow, Packman, & Harrison (in press). In the Lidcombe Program, the parent gives verbal contingencies during conversational exchanges with the child. These verbal contingencies are directed at: (1) stutter-free speech; (2) unambiguous stuttering; (3) correct self-evaluation of stutter-free speech; and (4) spontaneous self-correction of stuttering. These verbal contingencies consist of: (1) acknowledgment and/or praise for periods of stutter-free speech; (2) acknowledgment of stuttering and/or a request that the child corrects stuttering; (3) praise for correct self-evaluation of stutter-free speech; and (4) praise for spontaneous self-correction of stuttering. The SLP ensures that these parental verbal contingencies are not constant, intensive or invasive, and that parents are at all times positive and supportive of the child receiving the treatment. The treatment is individualized for each family and, as with any treatment for a childhood speech or language disorder, it is essential that the child enjoys the treatment and finds it to be a positive experience. Stuttering measures are an essential component of the Lidcombe Program. The parent makes daily measures of the severity of the child’s stuttering on a 10-point scale, where 1 = no stuttering; 2 = very mild stuttering; and 10 = extremely severe stuttering. The SLP makes weekly measures of stuttering rate (percent syllables stuttered, %SS). Together, these two measures are used to: (1) guide implementation of the program from week to week; (2) identify when the child has met criterion speech performance; and (3) check that the child’s speech continues to meet criterion speech performance in the long-term. The stuttering measures also enable the SLP and the parent to communicate effectively about the severity of the child’s stuttering throughout the treatment process. The Lidcombe Program is conducted in two stages. Stage 1 is complete when the child’s stuttering is below 1.0%SS and each of the daily severity ratings for the corresponding week are either 1 or 2, with the majority being 1. During Stage 2, the parent gradually withdraws the verbal contingencies and gradually assumes complete responsibility for the treatment as visits to the clinic decrease in frequency. Any departure from the criterion speech performance, as specified with the stuttering measures at the end of Stage 1, results in more frequent clinic visits and possibly an increase in parental contingencies.
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
205
A file audit of 250 children treated with the Lidcombe Program (Jones, Onslow, Harrison, & Packman, 2000) showed that the mean number of weekly clinic visits taken to complete Stage 1 was 12.5, and a recovery plot for these children showed that 90% of them had completed Stage 1 after 22 clinic visits. Jones et al. found that pre-treatment stuttering rate was the only predictor of time required to complete Stage 1. Other case variables — most notably time since reported onset — did not affect the time taken to complete Stage 1. The safety and clinical promise of the Lidcombe Program have been established in Phase I and II (Pocock, 1996) trials. Non-controlled trial data (Onslow, Andrews, & Lincoln, 1994; Onslow, Costa, & Rue, 1990) and file audit data (Jones, Onslow, Harrison, & Packman, 2000) have shown outcomes of zero or near zero rates of stuttering in children in the medium-term after treatment with the Lidcombe Program. Long-term data on 42 children (Lincoln & Onslow, 1997) show outcome to be durable. The social validity of treatment outcomes has been established (Lincoln, Onslow, & Reed, 1997). Given that the Lidcombe Program is a direct treatment, it is perhaps not surprising that concerns have been expressed about its safety (Cook, 1996; Cook & Rustin, 1997). Consequently, Woods, Shearsby, Onslow, and Burnham (2002) demonstrated that the treatment was not associated with psychological harm to children. To date, however, the efficacy of the Lidcombe Program has not been established. In other words, there is no unequivocal evidence that the reductions in stuttering that occur with the Lidcombe Program are in fact due to the treatment. The most important issue to consider in determining the effectiveness of treatments for early stuttering is, of course, natural recovery. Here, there are two questions (Packman & Onslow, 1998). The first is whether recovery from stuttering with the Lidcombe Program is greater than natural recovery in the long-term. In other words, after a clinically significant post-treatment period, are the improvements in stuttering that occur after treatment with the Lidcombe Program greater than those of natural recovery? The definitive answer to this question would involve comparing a group of children who receive no treatment with a group of children who receive treatment with the Lidcombe Program. In light of the fact that natural recovery may take 2 years or longer (e.g., Ingham & Riley, 1998; Yairi & Ambrose, 1999), this would involve withholding treatment for a control group of children for a number of years. This cannot be justified on ethical grounds, so the next best thing is to compare the Lidcombe Program with another treatment or treatments. Consequently, a Phase III (Pocock, 1996) randomized controlled trial of the program is in progress in New Zealand (Jones, Gebski, Onslow, & Packman, 2001). A second, and equally important question about effectiveness concerns the immediate effects of this treatment: When the Lidcombe Program is introduced, are its effects on the course of the disorder greater than the effects of natural recovery? Natural recovery occurs in many cases of early stuttering, with two recent estimates lying in the mid-70% range (Mansson, 2000; Yairi & Ambrose, 1999). At present, however, there is some doubt that recovery rates for children who present to a clinic would be as high as those population estimates. In any
206
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
event, in many instances SLPs will want to implement treatment sometime during the pre-school years, despite not knowing whether the child is destined to recover naturally. There are many reasons for clinicians to treat a pre-school child who stutters, even though there is a chance that natural recovery may occur at some future time (Packman & Onslow, 1999; Packman, Onslow, & Attanasio, in press). Consequently, there is need for evidence that the immediate reductions in stuttering observed with the Lidcombe Program can be attributed primarily to the treatment rather than to natural recovery. An experimental design is required to establish this evidence. Such a design involves withholding treatment for a control group, but only for a short period, and would determine whether the Lidcombe Program provides short-term benefits beyond what might be expected from natural recovery. The present paper reports such an experimental investigation. In this study, a group of children who receive the Lidcombe Program over 12 clinic visits is compared with a control group of children who receive no treatment for 12 weeks. Twelve clinic visits was chosen as this is close to the mean number of clinic visits (12.5) taken by pre-school children to complete Stage 1 of the Lidcombe Program (see earlier). Of course, not all children receiving the Lidcombe Program in the study would be expected to complete Stage 1 within 12 clinic visits, but this was considered a sufficient period to demonstrate the impact of the treatment on the natural course of stuttering.
2. Method 2.1. Participants The participants were 23 children assessed at the Stuttering Unit, Bankstown Health Service, and Campbelltown Hospital, Macarthur Health Service, in Sydney. The children lived in a range of suburbs in metropolitan Sydney of varying socio-economic levels. All children came to the clinic originally for stuttering, except for one child in the experimental group who presented with mild speech and language problems and who was found at assessment to be stuttering. At the conclusion of the experiment, one control child received additional treatment for language and voice problems, and one experimental child received additional treatment for articulation and language. The post-experiment treatment history of one experimental child and one control child was not available. Potential subjects were screened against selection criteria, which were: (1) age between 2 years 0 months and 4 years 11 months; (2) no previous treatment with the Lidcombe Program or professional consultation involving the Lidcombe Program; (3) 6 months or longer since reported onset of stuttering; (4) English the first language for both parents and child; and (5) stuttering at a rate of 3.0%SS or greater within the clinic during a conversation with the first investigator or a parent. For criterion (5), stuttering rate in %SS was obtained from an audio recording of a within-clinic speech sample recorded while the child played in the clinic at the
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
207
assessment interview. Potential subjects at the Stuttering Unit were recruited in the order in which parents telephoned to request treatment for their child. Potential subjects from Campbelltown Hospital were recruited in order from a list of children waiting for assessment. Potential subjects were assessed during an interview with the first investigator. 2.2. Procedure Parents of children who met selection criteria at the assessment interview were informed about the study and asked if they wished to participate. Parents were told that: (1) they would be randomly allocated to either an experimental (treatment) group or a control (non-treatment) group; (2) participation in the study would last approximately 12 weeks; and (3) at the completion of the study, children allocated to the treatment group would continue on the program and children in the control group would receive treatment when their name came to the top of the clinic waiting list. Parents who agreed to participate were supplied with a small audiotape recorder and instructions to make 10 min beyond-clinic recordings in three situations during the following week. In these situations the child was to speak: (1) with a parent; (2) with a familiar person other than a parent, away from home; and (3) at any location with a familiar person without the child’s knowledge of the recording. At the assessment interview, information about the nature of stuttering was provided to parents but no detailed information was provided about either the Lidcombe Program or any other stuttering treatment. Parents were advised only that the Lidcombe Program is presented by parents and involves weekly visits to the speech clinic. An appointment was made for the following week and parents were advised that their child would be randomly allocated to either the treatment or the control group when they returned at that time with all three tape recordings completed. No advice was given to any parent about how to react to or otherwise deal with the child’s stuttering. Children were randomly allocated to either the treatment group or the control group at the beginning of the first clinic visit after the interview assessment visit, provided parents had made and produced the three beyond-clinic recordings. Randomization was performed independently at the Clinical Trials Center of the National Health and Medical Research Council of Australia. The first investigator telephoned the Clinical Trials Center, where the child was randomized using a blocked randomization design. As this was an experiment and not a clinical trial, an a priori decision was made that randomization would continue until there were at least 10 subjects in each group, even if that resulted in uneven group numbers. Subject numbers were not based on power calculations, because, as is often the case with a low incidence disorder such as stuttering (Jones, Gebski, Onslow, & Packman, 2002), requisite numbers of clinical subjects for adequate power were prohibitive in terms of feasibility for an initial study. In the present case, greater than 60 subjects per group would be required to detect a medium effect at power
208
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
0.8 and α 0.05. Instead, subject numbers were chosen so that the present study could serve to determine the viability of a larger scale project in the event that data were non-significant in the presence of apparent trends (Jones, Gebski, Onslow, & Packman, 2002). Children allocated to the treatment group began their 12 clinic visits immediately. For these children, one within-clinic recording and three beyond-clinic recordings were again made, in the same speaking situations, after the requisite 12 clinic visits. The parents of the control group children were scheduled to return to the clinic 12 weeks post-randomization. During this visit, a within-clinic recording was made, and the parents presented the required set of three beyond-clinic recordings made during the previous week. 2.3. Dependent measure The dependent measure was %SS. In contrast to definition-based measures of stuttering, the measure of %SS records the number of speech events thought by an experienced clinician to be unambiguous stuttering (Jones, Gebski, Onslow, & Packman, 2001). The %SS does not include counts of normal disfluencies. Measures of %SS were made by an SLP experienced in treating and measuring stuttering, who was independent of the study and had no knowledge of the participants. The clinician knew the topic of the research but not its details. Recordings were presented blind, in random order, to this clinician with pre-treatment and post-treatment status also randomized. The mean %SS scores for each child at 1 week pre-randomization (first measurement occasion) and at 12 weeks post-randomization (second measurement occasion) were calculated, based on the four tape recordings (three made beyond the clinic and one made within the clinic). Pre- and post-randomization within-clinic recordings for all children were selected for intrajudge reliability analysis. These 46 recordings, comprising around a quarter of the total recordings, were presented to the SLP for re-measurement, 1 month after the original measures were made. Interjudge reliability was established by a second SLP, who was also experienced in treating and measuring stuttering, and who was also independent of the study and had no knowledge of the participants. This clinician also knew the topic of the research but not its details. This SLP also made blind measurements from randomly arranged recordings. It was determined a priori that around one-third of the recordings would be included in the interjudge reliability analysis. Tapes were selected randomly from the pool of recordings and presented to the SLP for measurement.
3. Results Twenty-nine children were randomized to the experiment. Of these, six children did not complete the experimental procedures, for the following reasons: death in
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
209
Table 1 Number, gender, mean %SS and mean time since onset for the control and the treatment groups Number
Control Treatment
Boys
Girls
11 8
2 2
Mean %SS pre-randomization
Mean time since onset (months)
8.4 8.6
10 12
the family, failure to comply with experimental procedures, treatment sought closer to home and loss of contact with the family. This left 10 children in the treatment group and 13 in the control group. Information about the children is shown in Tables 1 and 2.
Table 2 The columns display, for individual children, gender, reason for individual referral, treatment subsequent to the experiment for communication disorders other than stuttering, months since reported onset, and reported history of stuttering in first- or second-degree relatives Subject
Group
Gender
Original referral
Subsequent treatment
Months since reported onset
Reported family history
1 2
Control Control
M M
Stuttering Stuttering
9 12
Unknown Yes
3 4 5 6 7 8 9 10 11
Control Control Control Control Control Control Control Control Control
M M M F M M M M M
Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering
10 6 12 12 12 12 6 12 12
Yes Unknown Unknown No Yes Yes Yes Yes Yes
12 13 14 15
Control Control Experimental Experimental
F M M M
Stuttering Stuttering Stuttering Stuttering
6 9 12 12
Yes Yes Yes Yes
16
Experimental
M
6
Yes
17 18 19 20 21 22 23
Experimental Experimental Experimental Experimental Experimental Experimental Experimental
F M M M M M F
Mild articulation/ language Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering Stuttering
No Language/ voice No No No No No No No No History unavailable No No No History unavailable Language
16 11 12 8 12 13 18
Yes Yes Yes Yes No Yes Yes
No No No No No No No
210
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
Of the 184 recordings scheduled in the experiment (23 children×4 recordings× 2 assessment occasions), 178 (96.7%) were obtained by the investigators and used for data collection. The mean duration of tape recordings of the children was 9.2 min. The mean number of syllables counted on the recordings during the collection of %SS scores was 448 (S.D. = 159; range, 93–959). In order to avoid multiple statistical comparisons within a small data set, and in order to minimize the effects of the variability of early stuttering on analysis procedures, the stuttering rates for each child pre- and post-randomization were based on means of four recordings. Therefore, mean %SS scores in the analysis were generally based on speech samples in excess of 1500 syllables. Of the 46 samples re-measured for intrajudge reliability, 37 (78.7%) differed by less than 1.0%SS, 46 (97.9%) differed by less than 2.0%SS, and 46 (100%) differed by less than 3.0%SS. The Pearson correlation between the original ratings and these ratings was 0.99. Of the 66 samples re-measured by the independent clinician for interjudge reliability analysis, 18 (27.3%) differed by less than 1.0%SS, 39 (59.1%) differed by less than 2.0%SS, and 55 (83.3%) differed by less than 3.0%SS. The Pearson correlation between the original ratings and these ratings was 0.83. The results of intrajudge and interjudge reliability analyses showed that overall measurement error for %SS was likely to be small in relation to the data trends detected in the study (see Fig. 1), and that data trends pre- to post-randomization would be detectable with the %SS measure. The mean %SS scores for the treatment and the control groups at the pre-randomization measurement occasion were 8.6 (S.D. = 5.2; range, 2.6–20.1) and 8.4 (S.D. = 2.5; range, 3.6–13.1) respectively, and at the post-randomization measurement occasions were 3.5 (S.D. = 2.8; range, 0.6–9.2) and 5.8 (S.D. = 3.6; range, 2.3–15.3), respectively. Nine of the 10 experimental children reduced their
Fig. 1. Mean %SS scores for the treatment and control group pre-randomization and postrandomization.
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
211
%SS score from pre- to post-randomization, with the mean change in those children being a 39% reduction of the pre-randomization %SS score. The remaining participant’s pre-randomization score increased by 16%. Nine of the 13 control children reduced their %SS score from pre- to post-randomization, with the mean change in those nine children being a 26% reduction of the pre-randomization %SS score. The other four control children increased their stuttering rates by 54, 20, 42, and 6% of the pre-randomization score. As expected, %SS data were skewed (Jones, Onslow, Harrison, & Packman, 2000). Normally distributed data are necessary for the calculation of variances in ANOVA. Hence, data were log transformed to produce a normal looking distribution and analyzed with a repeated measures ANOVA. There was no main effect for group (control/treatment; F = 3.13, P > 0.05). In other words, there was no difference between the groups when both pre- and post-randomization %SS scores were taken into account. There was a significant main effect for measurement occasion (pre- and post-randomization; F = 28.32, P < 0.001). This means that when %SS scores for both groups are combined, there was a significant decrease in stuttering from the first to the second measurement occasion. There was a significant interaction between group and measurement occasion (F = 5.02, P < 0.05). In other words, the treatment group improved significantly more than the control group from the first to the second measurement occasion. The significant interaction is displayed in Fig. 1.
4. Discussion The long-term effects of the Lidcombe Program are being studied in the Jones, Gebski, Onslow, & Packman (2001) clinical trial. The present study dealt with short-term effects, and showed that the improvement in stuttering in a group of pre-school-age children after a set number of clinic visits in the Lidcombe Program was greater than the effect of natural recovery. In other words, the results indicate that the treatment has an immediate impact on the natural course of the disorder. While some reduction in %SS scores occurred in the children who did not receive treatment, the decrease in stuttering in the treated group was around twice that amount. These results show that clinicians can be confident that implementing the Lidcombe Program will have a positive impact on stuttering rate in their pre-school clients. Stuttering frequency was the only dependent variable in the present study, so these results offer no explanation for why the treatment might be effective. No attempt was made to investigate reasons for the improvement in either group in this study, because measures of stuttering were made only after 12 weeks and before the completion of treatment, which is not a suitable design for such an investigation. Nonetheless, the reasons for recovery from stuttering during the Lidcombe Program is an interesting issue. At present, preliminary data suggest that neither extensive change in children and parents’ language habits (Bonelli, Dixon,
212
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
Bernstein Ratner, & Onslow, 2000), nor changes in speech acoustics (Onslow, Hewat, McLeod, & Packman, 2002), are associated with the treatment. The use of the present controlled experimental method would be an effective means for continuing this search for any changes in children or their parents that might explain the effects of the Lidcombe Program. The design of this experiment involved randomizing children, as they entered the study, either to the treatment group or the control group. This design, which is a true experimental design, was chosen in preference to a matched cohort design, which is a quasi-experimental design. The experimental design was chosen because it eliminates bias. It might be argued that children in a study such as the present one should be matched on variables such as stuttering rate, age, time since onset of stuttering, gender, and family history of recovery from stuttering. It could also be argued that they should be matched for cognitive, language and phonological development. Matching on so many variables would be unmanageable, and failure to match on any or all of them would leave the design open to the effects of bias. Even if matching on all such variables were to be achieved, there may be other variables, yet unknown to us, that might bias the results of such a study. The control children, as a group, showed some improvement in stuttering during the period of study, yet it is difficult to interpret this result considering that these are the first reported clinical measures of %SS over so short a period for a group of stuttering pre-school children who do not receive treatment. It may well be the case that some of the children in the control group who showed clinically significant stuttering reductions during the period of study may have continued that trend and recovered from the condition in a timely fashion. The present results raise the need to explore this issue in further clinical investigations. References Bonelli, P., Dixon, M., Bernstein Ratner, N., & Onslow, M. (2000). Child and parent speech and language and the Lidcombe Program of early stuttering intervention. Clinical Linguistics and Phonetics, 14, 427–446. Cook, F. (1996). The Lidcombe Programme: Is this the cure? Bulletin of the Royal College of Speech and Language Therapists, 528, 14. Cook, F., & Rustin, L. (1997). Commentary on the Lidcombe Program of early stuttering intervention. European Journal of Disorders of Communication, 32, 250–258. Ingham, J. C., & Riley, G. (1998). Guidelines for documentation of treatment efficacy for young children who stutter. Journal of Speech, Language, and Hearing Research, 41, 753–770. Jones, M., Gebski, V., Onslow, M., & Packman, A. (2001). Design of randomized controlled trials: Principles and methods applied to a treatment for early stuttering. Journal of Fluency Disorders, 26, 1–21. Jones, M., Gebski, V., Onslow, M., & Packman, A. (2002). Power in health research: A tutorial. Journal of Speech, Language, and Hearing Research, 45, 243–255. Jones, M., Onslow, M., Harrison, E., & Packman, A. (2000). Treating stuttering in children: Predicting treatment time in the Lidcombe Program. Journal of Speech, Language, and Hearing Research, 43, 1440–1450. Lincoln, M., & Onslow, M. (1997). Long-term outcome of an early intervention for stuttering. American Journal of Speech–Language Pathology, 6, 51–58.
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
213
Lincoln, M., Onslow, M., & Reed, V. (1997). Social validity of an early intervention for stuttering: The Lidcombe Program. American Journal of Speech–Language Pathology, 6, 77–84. Mansson, H. (2000). Childhood stuttering: Incidence and development. Journal of Fluency Disorders, 25, 47–57. Onslow, M., Andrews, C., & Lincoln, M. (1994). A control/experimental trial of an operant treatment for early stuttering. Journal of Speech and Hearing Research, 37, 1244–1259. Onslow, M., Costa, L., & Rue, S. (1990). Direct early intervention with stuttering: Some preliminary data. Journal of Speech and Hearing Disorders, 55, 405–416. Onslow, M., Hewat, S., McLeod, S., & Packman, A. (2002). Speech segment timing in children after the Lidcombe Program of early stuttering intervention. Clinical Linguistics and Phonetics, 16, 21–33. Onslow, M., Packman, A., & Harrison, E. (in press). The Lidcombe Program of early stuttering intervention: A clinician’s guide. Austin, TX: Pro-Ed. Packman, A., & Onslow, M. (1998). What is the take-home message from Curlee and Yairi? American Journal of Speech–Language Pathology, 7, 5–9. Packman, A., & Onslow, M. (1999). Issues in early intervention. In M. Onslow & A. Packman (Eds.), Handbook of early stuttering intervention. San Diego, CA: Singular Publishing Group. Packman, A., Onslow, M., & Attanasio, J. (in press). The timing of intervention with the Lidcombe Program. In M. Onslow, A. Packman, & E. Harrison (Eds.), The Lidcombe Program of early intervention: A clinician’s guide. Austin, TX: Pro-Ed. Pocock, S. J. (1996). Clinical trials: A practical approach. Chichester, England: Wiley. Woods, S., Shearsby, J., Onslow, M., & Burnham, D. (2002). The psychological impact of the Lidcombe Program of early stuttering intervention: Eight case studies. International Journal of Language and Communication Disorders, 37, 31–40. Yairi, E., & Ambrose, N. G. (1999). Early childhood stuttering I: Persistency and recovery rates. Journal of Speech, Language, and Hearing Research, 42, 1097–1112.
CONTINUING EDUCATION An experimental investigation of the impact of the Lidcombe Program on early stuttering QUESTIONS 1. This experiment was designed to determine: a. the long-term effects of the Lidcombe Program on stuttering b. the short-term effects of the Lidcombe Program on stuttering c. the long-term effects of natural recovery on stuttering d. the short-term effects of natural recovery on stuttering e. the relative effects of the Lidcombe Program and natural recovery on stuttering 2. The Lidcombe Program of early stuttering intervention is: a. a speech-pattern change procedure b. a procedure where demands on the child are reduced c. a behavioral treatment d. a cognitive-behavior treatment e. all of the above 3. In this experiment, the dependent variable was: a. the group to which the children were assigned b. percentage of syllables stuttered
214
V. Harris et al. / Journal of Fluency Disorders 27 (2002) 203–214
c. stuttering severity measured on a 10-point scale d. syllables per minute e. “b” and “d” above 4. The design of this experiment was: a. a randomized controlled design b. a quasi-experimental design c. a crossover design d. a double-blind design e. none of the above 5. In this experiment, children in the treatment group reduced their stuttering: a. almost to the same extent as the control children b. twice as much as the control children c. half as much as the control children d. more than any other treatment group studied e. more than one of the above